Goldman's NFP Take: "Much Weaker Than Headlines Suggest"

From Goldman's Jan Hatzius

USA: Non-Farm Payroll Employment - Much weaker than headlines suggest
   
Actual: -125,000 mom
Previous: 433,000 mom
Consensus: -130,000
Released: Friday, July 02, 2010 at 08:30 (New York time)

Much weaker than headlines suggest

BOTTOM LINE: The headlines of the June employment report were decent - a rise in private sector employment by 83,000 and a drop in the unemployment rate from 9.7% to 9.5% - but the "guts" were much weaker as household employment declined sharply and both the workweek and average hourly earnings declined. The report confirms that the labor market has not yet reached the self-sustaining recovery stage, at a time when forward-looking indicators of economic activity are slackening. We have judgmentally downgraded the US-MAP rating to -2 from the 0 suggested by the inline "headline" payrolls reading because of the much weaker guts of this report.

US-MAP:
Nonfarm Payrolls -10 (5, -2).

KEY NUMBERS:
Nonfarm payrolls -125k in June vs. GS -100k, median forecast -130k.
Private payrolls +83k in June vs. GS +150k, median forecast +110k.
Unemployment rate 9.5% in June vs. GS and median forecast 9.8%.
Average hourly earnings -0.1% in June (mom, +1.7% yoy) vs. GS and median forecast +0.1%.

MAIN POINTS:

1. Nonfarm payrolls fell 125,000, more than accounted for by a 225,000 drop in Census employment. Elsewhere, the private sector added 83,000 jobs and non-Census government employers 17,000. These numbers are fairly close to consensus expectations, especially once we take into account a cumulative net upward revision of 25,000 to total payrolls in May. There is some evidence that private sector hiring benefited from the unwind of a Census "displacement" effect in May, both in the pickup in overall private employment growth (83k vs. 33k in May) and in the sectoral distribution of gains, with a sharp rebound in leisure and hospitality employment (+37k after -7k in May). The establishment survey benefited from a birth-death adjustment of +147k, 14k more than in June 2009.

2. Other aspects of the establishment survey were considerably weaker. The payroll diffusion fell to 52.2 in the nonfarm business sector, from 54.8 in May and a recent high of 68.0 in April. The workweek dropped 0.1 hours to 34.1, with almost all of this drop due to a sharp 0.5-hour drop in manufacturing. Moreover, average hourly earnings fell 0.1%, on top of a downward revision to the prior month. Average hourly earnings now stand at just+ 1.7% year-on-year, a new cycle low. Hence, the implications for private-sector income generation as well as inflation are negative.

3. The household survey was very weak as well. The drop in the unemployment rate was entirely due to a huge 652k drop in the labor force, undoubtedly partly due to the dropout of Census workers from the labor force. Only some of these workers seem to have gone into the "pool of available workers" as broader measures of underemployment were mostly down as well. However, the Census cannot explain the fact that the employment/population ratio fell back to 58.5% in June from a peak of 58.8% in April (when the level of Census workers was lower than in June).